I figured a first step, and sadly non-AI related would be an introduction. No, I am not an AI myself, I am a human. My Name is Matthew Koster, and I also run another blog called “Considered Normal“, which focuses mostly on tech tutorials and my life on the Island (Montreal). But as I have moved back home to Ontario, I figured I would take on a new project.

AI has always intrigued me, and none more than those that generate images. Now there is a lot of controversy over the topic of AI Art, and I am sure some of it will be covered here at ArtiVerse, but I see Generative AI as a tool. I come from a very talented family, my Father was a graphic designer, my mother an artisan, and my brother is a formally trained artist (graduate of OCA). Although not formally trained in anything remotely artistic, I was gifted with an eye for it. I could tell you that certain colors look good together, however, my brother could tell you exactly why and all the theory behind it. But other than that, I have no artistic skill whatsoever outside my knowledge of photography and the use of Photoshop. So to me, tools like MidJourney, Automatic1111, ComfyUI, and Fooocus are how I can get my ideas down on paper. Was I the artist who created the work of art? No. But it was my wordsmithing that choreographed the outcome.

MidJourney is where I started, its simple interface made it very easy to generate some cool-looking images. But for what I wanted it for, it would cost as to reuse the images commercially, you need a paid package. The fact it was built into Discord though was very appealing, as I was able to create a private channel on my Discord server (with the paid version) and keep my generations semi-private (they would not show up in the general chat channels). But it was not fully private, my image generations still went to the community page where they could be viewed.

Although this tool is VERY powerful and requires little to no skill to use, I wanted to dive deeper. In comes Stable Diffusion, in the form of Automatic1111. This meant I could use my computer system to generate the images, which meant that they would be 100% private and not visible anywhere else on the net until I decided to upload them. Many of these designs you can see over at my gallery on Considered Normal (https://considerednormal.com/gallery/).

From there I got into Comfy, which is an amazing tool that quite reminds me of no-code programming and paths the like you would find in Blender. I watched a lot of videos on how to create some good workflows, and this was about the time that the Stable Diffusion XL (SDXL) checkpoint was released.

Although I was fairly happy with the workflow I created, It was still very hard to get the images I wanted. Both A1111 and ComfyUI have a learning curve, Comfy more so.

Most recently I came across Fooocus in a youtube video, it simplifies a lot of the tasks needed to get good images in A1111, and ComfyUI its more akin to MidJourney than any of the other tools I have seen, although not as intuitive and easy to use as MidJourney, it is fairly simple to get some good quality photos out of it.

So this is where I am at. I am nowhere near an expert on the topic of AI, and I have a lot I want to learn. But it just fascinates me that all the images (other than the workflow) in this post were generated by AI based on the words I gave it. It just blows my mind…

Now, as for my chosen author name… I go by many names on the net, but for this journey, I have chosen the moniker Artifice Prime, which is sort of a poke at AI:

ar·ti·fice – clever or cunning devices or expedients, especially as used to trick or deceive others.

Oxford Languages and Google

As for Prime, well, who isn’t a fan of the Autobots?

I hope you join me on my journey into the ARTIficial UniVERSE